The objective of the proposed research is the development of a general and practicable statistical framework and associated novel methodology for the analysis of genome-wide association studies. This framework will allow us to address many open design and analysis questions in a coherent fashion. The specific aims of our proposal are the development of: efficient experimental designs for genome-wide association studies; novel and powerful statistical methods for genome-wide multi-marker association testing; methodology for the detection of interacting disease genes; and analysis tools for the joint analysis of multiple related genome wide scans. Key tools in this work will be the use of coalescent and population genetics approaches, incorporation of newly available fine-scale genetic maps for the human genome, and the exploitation of recent developments in computationally intensive statistical methods, notably in a missing data framework. We plan to integrate existing and new methodology into a comprehensive suite of analysis software which will be made publicly available. We envisage working closely on all aspects of this project with other research groups funded as part of the proposed consortium. We would bring to this consortium the data from the Wellcome Trust Case Control Consortium, a major UK-based genome-wide association study of 2000 cases and 3000 controls for each of eight common human diseases, together with considerable experience with that data and that from the International HapMap project. The research outlined in this proposal will lead to efficient and practicable studies of human genetic disease. The ultimate goal of such studies is the development of individual disease prevention and treatment, novel therapeutics, and maintenance of health. [unreadable] [unreadable] [unreadable]